1 / 30

Indian summer monsoon rainfall

Indian summer monsoon rainfall in the ECMWF seasonal fc. System-3: predictability and ISV diagnostics. Franco Molteni European Centre for Medium-Range Weather Forecasts, Reading, U.K. contributions from: ECMWF Seasonal Forecast Section, Laura Ferranti, Milind Mujumdar (IITM, Pune).

mrosario
Download Presentation

Indian summer monsoon rainfall

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Indian summer monsoon rainfall in the ECMWF seasonal fc. System-3: predictability and ISV diagnostics Franco Molteni European Centre for Medium-Range Weather Forecasts, Reading, U.K. contributions from: ECMWF Seasonal Forecast Section, Laura Ferranti, Milind Mujumdar (IITM, Pune)

  2. ECMWF Seasonal forecast system (Sys-3) IFS 31R1 1.1 deg. 62 levels HOPE 1.4 deg. lon 1.4/0.3 d. lat. OASIS-2 TESSEL Ens. Forecasts Initial Con. 4-D variational d.a. Gen. of Perturb. System-3 CGCM Multivar. O.I.

  3. The seasonal forecast System-3(implem. March 07) • COUPLED MODEL (IFS + OASIS2 + HOPE) • Recent cycle of atmospheric model (Cy31R1) • Atmospheric resolution TL159 and 62 levels • Time varying greenhouse gasses. • Includes ocean currents in wave model • INITIALIZATION • Includes bias correction in ocean assimilation. • Includes assimilation of salinity and altimeter data. • ERA-40 data used to initialize ocean and atmosphere in hindcasts • Ocean reanalysis back to 1959, using ENACT/ENSEMBLES ocean data • ENSEMBLE GENERATION • Extended range of back integrations: 11 members, 1981-2005. • Revised wind and SST perturbations. • Use EPS Singular Vector perturbations in atmospheric initial conditions. • Forecasts extended to 7 months (to 13 months 4x per year).

  4. Products from Sys-3: “plumes” for El Nino indices

  5. Products from Sys-3:’tercile summary’

  6. Climagrams : area-averages of 2mT and rainfall

  7. Climagrams :monsoon indices / teleconnections

  8. Climagrams :SST area-averages

  9. All India Rainfall: “climagram” from 1 May 2007

  10. Verification for SST in JJA from 25-year hindcast set

  11. Climagram verification (ACC : f. of initial and lead time) SE Trop. Ind. Oc. Nino3.4 S Trop. Atlantic

  12. Anomaly correlation of seasonal-mean rainfall

  13. Predictability of teleconnection/EOF indices in S-3 Rainfall: East. Tropical Indian Ocean pattern (JJA)

  14. Predictability of teleconnection/EOF indices in S-3 Rainfall: Sahel / Guinea coast dipole (JJA)

  15. Predictability of AIR in S-3 JJAS CC = .25 JAS CC = .46

  16. Monsoon onset in India: seasonal fc. for June 1-month-lead fc. 0-lead fc. cc = 0.35 cc = 0.45 Forecast anomaly amplitude is ~ 2 x obs. !

  17. Predictability of AIR in S-3: EOF filtered JAS Unfiltered CC = 0.46 EOF proj. CC = 0.59 CC = .50

  18. Predictability of East Africa short rains: EOF filtered OND Unfiltered CC = 0.04 EOF proj. CC = 0.39

  19. All-India Rainfall time-series (May-September) 2002 2003 2004

  20. Climagram verification (ACC : f. of initial and lead time) All-India Rainfall

  21. Diagnostics of monsoon ISV in Sys3 from pentad data • Observations (5-day means from 2 June to 30 Sept. 1981-2005 ) • Rainfall: 5-day means computed from India Met. Dept. daily rainfall analysis on 1x1 deg grid (Rajeevan et al. 2005) • 850-hPa relative vorticity from ERA40 and operational ECMWF analysis (2002 onwards) • Model data: • 5-day means of rainfall and 850-hPa vorticity from seasonal hindcast ensembles started on 1 May 1981-2005 • Diagnostics: • EOFs of pentad rainfall over India (70-90E, 7.5-30N) in July-August • Covariance of 850-hPa vorticity with 1st PC of rainfall at lag 0, -1, -2, -3 pentads

  22. Rainfall mean, st.dev. and EOF domain for July-August IMD data Sys3

  23. 5-d mean rainfall EOFs for July-August 1981-2005 IMD data Sys3

  24. 5-d mean rainfall EOFs for July-August IMD data 1951- -2007 Sys3 1981- -2005

  25. Vorticity 850 hPa: stand.dev. and cov. with rainfall PC1 Era40 + oper. An. Sys3

  26. Vorticity 850: lagged cov. with rainfall PC1 (ERA40) Vor (t-1) Vor (t-3) Vor (t0) Vor (t-2)

  27. Vorticity 850: lagged cov. with rainfall PC1 (Sys3) Vor (t-1) Vor (t-3) Vor (t0) Vor (t-2)

  28. 5-d mean rainfall EOFs for July-August IMD data 1951- -2007 Observed Rainfall June-July 2008

  29. Conclusions (1) • SST predictions from the ECMWF seasonal forecast system-3 show higher skill than those from previous system, particularly in the tropical Pacific and eastern Indian Ocean, but western Indian Oc. and tropical Atlantic are still not better than persistence in NH summer.. • Predictive skill for seasonal rainfall is generally good over the Pacific and tropical S. America, poor along the coast of the Indian Ocean in early summer. Skill for South Asian rainfall increases in the latter part of the monsoon season. • Seasonal forecasts over land can be improved by exploiting teleconnections with adjacent ocean regions.

  30. Conclusions (2) • Predictive skill for monthly means of Indian rainfall in seasonal forecasts is generally low, with a maximum for September . • 5-day mean rainfall analysis shows a deficient representation of variability in the northern part of India, while variability in the central peninsula is overestimated by the Sys3 model. This is reflected in the standard deviation of 850-hPa relative vorticity. • Lag covariance maps indicate a less coherent organization of convection in the Sys3 model with respect to observation, with less evidence of meridional propagation and weaker teleconnections between the Indian and the South Asian region. • Statistical corrections may be needed to extract information about year-to-year variations of monsoon ISV from Sys3 output.

More Related